Learning to recognize video-based spatiotemporal events

Harini Veeraraghavan, Nikolaos P Papanikolopoulos

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

A key research issue in activity recognition in real-world applications, such as in intelligent transportation systems (ITS), is to automatically learn robust models of activities that require minimal human training. In this paper, we contribute a novel approach for learning sequenced spatiotemporal activities in outdoor traffic intersections. Concretely, by representing the activities as sequences of actions, we contribute a semisupervised learning algorithm that learns activities as complete stochastic context-free grammars (SCFGs), namely, the grammar structure and the parameters. Our approach has been implemented and tested on real-world scenes, and we present experimental results of the grammar learning and activity recognition applied to datacollection and traffic monitoring applications using video data.

Original languageEnglish (US)
Article number5166486
Pages (from-to)628-638
Number of pages11
JournalIEEE Transactions on Intelligent Transportation Systems
Volume10
Issue number4
DOIs
StatePublished - Dec 1 2009

Fingerprint

Context free grammars
Learning algorithms
Monitoring

Keywords

  • Context-free grammars
  • Intelligent transportation system (ITS) applications
  • Machine learning
  • Vehicle tracking
  • Video analysis

Cite this

Learning to recognize video-based spatiotemporal events. / Veeraraghavan, Harini; Papanikolopoulos, Nikolaos P.

In: IEEE Transactions on Intelligent Transportation Systems, Vol. 10, No. 4, 5166486, 01.12.2009, p. 628-638.

Research output: Contribution to journalArticle

@article{a10ef7a976b34d93bafdfacac3cefe3e,
title = "Learning to recognize video-based spatiotemporal events",
abstract = "A key research issue in activity recognition in real-world applications, such as in intelligent transportation systems (ITS), is to automatically learn robust models of activities that require minimal human training. In this paper, we contribute a novel approach for learning sequenced spatiotemporal activities in outdoor traffic intersections. Concretely, by representing the activities as sequences of actions, we contribute a semisupervised learning algorithm that learns activities as complete stochastic context-free grammars (SCFGs), namely, the grammar structure and the parameters. Our approach has been implemented and tested on real-world scenes, and we present experimental results of the grammar learning and activity recognition applied to datacollection and traffic monitoring applications using video data.",
keywords = "Context-free grammars, Intelligent transportation system (ITS) applications, Machine learning, Vehicle tracking, Video analysis",
author = "Harini Veeraraghavan and Papanikolopoulos, {Nikolaos P}",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/TITS.2009.2026440",
language = "English (US)",
volume = "10",
pages = "628--638",
journal = "IEEE Intelligent Transportation Systems Magazine",
issn = "1524-9050",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Learning to recognize video-based spatiotemporal events

AU - Veeraraghavan, Harini

AU - Papanikolopoulos, Nikolaos P

PY - 2009/12/1

Y1 - 2009/12/1

N2 - A key research issue in activity recognition in real-world applications, such as in intelligent transportation systems (ITS), is to automatically learn robust models of activities that require minimal human training. In this paper, we contribute a novel approach for learning sequenced spatiotemporal activities in outdoor traffic intersections. Concretely, by representing the activities as sequences of actions, we contribute a semisupervised learning algorithm that learns activities as complete stochastic context-free grammars (SCFGs), namely, the grammar structure and the parameters. Our approach has been implemented and tested on real-world scenes, and we present experimental results of the grammar learning and activity recognition applied to datacollection and traffic monitoring applications using video data.

AB - A key research issue in activity recognition in real-world applications, such as in intelligent transportation systems (ITS), is to automatically learn robust models of activities that require minimal human training. In this paper, we contribute a novel approach for learning sequenced spatiotemporal activities in outdoor traffic intersections. Concretely, by representing the activities as sequences of actions, we contribute a semisupervised learning algorithm that learns activities as complete stochastic context-free grammars (SCFGs), namely, the grammar structure and the parameters. Our approach has been implemented and tested on real-world scenes, and we present experimental results of the grammar learning and activity recognition applied to datacollection and traffic monitoring applications using video data.

KW - Context-free grammars

KW - Intelligent transportation system (ITS) applications

KW - Machine learning

KW - Vehicle tracking

KW - Video analysis

UR - http://www.scopus.com/inward/record.url?scp=72649096732&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72649096732&partnerID=8YFLogxK

U2 - 10.1109/TITS.2009.2026440

DO - 10.1109/TITS.2009.2026440

M3 - Article

AN - SCOPUS:72649096732

VL - 10

SP - 628

EP - 638

JO - IEEE Intelligent Transportation Systems Magazine

JF - IEEE Intelligent Transportation Systems Magazine

SN - 1524-9050

IS - 4

M1 - 5166486

ER -